In a digital printing workflow there is a need to be able to proof bitmap files used to make printing plates. Presently, customer artwork consisting of contone images, linework, and text, is first sent to a digital halftone proofer or inkjet printer. The artwork is corrected until the proof is approved for the press. In the case were the artwork is proofed on a digital halftone proofer such as described by Baek et al. in U.S. Pat. No. 5,164,742, the raster image processor (RIP) adjusts the input continuous tone data using a calibration dot-gain curve such that the tone-scale of the proof matches the tone-scale of the press-sheet. After the proof is approved, the job is sent to a second RIP which applies a second dot-gain curve for generating the plate used in the press-run.
The first and second RIPs may be the same but are typically separate and may be located apart from each other. The first and second RIPs are preferably the same type and version such that the halftone dots created and algorithms used by each device are an exact match. Many times the two RIPs are not an exact match, which can create problems. Sometimes incorrect dot-gain correction files are used. Sometimes the artwork is changed between creating the proof and the plates and the press-run no longer matches the approved proof.
Another disadvantage in the current system is that an error in the creation of the bitmaps for printing is not known until the plates are loaded onto the press and the press-run is started. For a press capable of over 1,000 impressions per hour considerable amount of production is lost if the plates are found to be corrupt and need to be remade.
An important aspect in creating a halftone proof is predicting dot-gain or tone-scale. Dot-gain is a known phenomenon attributable to ink spread, ink absorption by the print media, and optical effects between the ink and the paper. The dot-gain varies with the size and shape of the halftone dots, the printing device, the inks, and the paper used, etc. For a digital proof, halftone dots in a color separation are composed of micro-pixels that give the halftone dot its shape and size. Dot-gain for a digital proof corresponds to increasing dot size by adding micro-pixels. Dot-loss for a digital proof corresponds to decreasing dot size by eliminating micro-pixels. Dot-gain correction consists of adding and subtracting gain to match the response at different percent dot inputs.
In the printer described by Baek et al. many steps are required to match the press. First, the exposure for each color plane is adjusted to match the solid area density. Second, the dot-gain for each color plane is adjusted to achieve a dot-gain match at different halftone tint levels. Third, the dot-gain curves and density levels may be fine tuned to achieve either a good neutral match in the three color overprints or a color match for flesh tones. For some work, other memory colors such as green grass or light blue sky may be matched as the critical color. Finally, the dot-gain curves may be further adjusted to deliver better performance in the highlight, or shadow areas. These steps are critical and typically take much iteration between the proof operator and the customer to achieve the look that the customer desires. It is important to be able to adjust the proofer to achieve this look as there are other controls on the press that may be adjusted to affect the dot-gain and tonal control of the press-run. By adjusting the performance of the proofer, the customer is selecting the quality of the proofs that will be used by the pressmen to match.
Once the proofer has been setup to match the press, the customer uses subsequent proofs to setup the press. This is an important point. The proofer setup is used to simulate the press such that the pressman may then use the proofs to setup the press to achieve the customer's intent. Every job going through the proofer will be adjusted with a setup. There may be different setups for each press or press type. There may also be different setups for different customers using the same proofer. Finally there may also be standard setups that are used to simulate jobs across many different presses.
The same job is typically “ripped” again when going to press. This time the RIP is programmed to generate 50% area coverage on plate for the 50% color input. The press is then run to deliver a fixed amount of gain at the 50% input level. Dot-gain is due to the smearing of the ink from the plate to a blanket, the smearing of ink from the blanket to the job paper, and the optical gain of the ink on top of the paper. The control is usually split between the plate making device delivering 50% area coverage for a 50% input, and the press delivering 50% plus its intrinsic dot-gain. Typical dot-gain levels for a Web-fed offset press are 15% to 25% at the 50% input level. Because the dot-gain occurs on the press instead of at the plate writer the bitmaps used to create the plate will not contain enough gain to make the proof. Proofs made from these bitmaps will be washed out and the contrast will be significantly reduced. Colors will also shift, as the gain in each color will be proportional to the dot area coverage.
Other digital halftone printing devices such as that disclosed by Michalson in U.S. Pat. No. 6,204,874 use a binary proofing media that does not allow for adjusting the density level of the solid colorants. A different process is used to adjust these devices for a close press match, including adjusting the tone-scale or dot-gain curve used to make the bitmap file. However, the ideal dot-gain curve on these systems is still different from the dot-gain curves used to make the plates even if the same machine is imaging the plate and the proof as disclosed by Michalson.
Inkjet printing devices are also sometimes used to make a proof. These devices typically image from 300 dpi to 1440 dpi writing resolutions using multiple cyan, magenta, yellow, and sometimes black inks. In addition software such as “Best Screen Proof” available from Best Gmbh, or Black Magic available from Serendipity Software Pty Ltd., may be used to simulate the printing of a halftone screen. This software attempts to measure the halftone screen and adjust the printed output to achieve a close color match to a given target. Resolution of the inkjet devices does not allow for a good match of the halftone dot structure. The color match developed does simulate the tone-scale or dot-gain correction, but only through the driving of the overlapping colors on the proof. The quality of the halftone in the printed proof is significantly compromised. Dots in the highlight and shadow areas are destroyed in trying to match the overall density level in these systems. This is because the inkjet output drops are too large. Therefore one inkjet drop is used to replace many halftone dots in the highlight or bright areas, while one inkjet hole is used to replace many halftone holes in the shadows.
A halftone screen at 150 lines per inch, 6 lines per mm, covers an area of approximately 28,674 μm2. An inkjet printer with a 3 pL drop size will produce a dot with a diameter of about 25 μm covering an area of 625 μm2. This may vary depending upon the spread into the paper. A single inkjet drop represents a 2.18% change in area within a 150 line screen halftone. To achieve finer resolution the Best Screen Proof, and Black Magic, software use additional inks to image multi-level colorants. Typically a light cyan and light magenta ink are added to the cyan, magenta, yellow, and black primaries to achieve finer control of the tone-scale. While this creates a proof with a close visual color match, the structure of the halftone dots within the image is seriously degraded.
The conventional proofing solution is to RIP the file for proofing separate from ripping the file for printing, adding dot-gain to the proofing file as part of the ripping process. U.S. Pat. No. 5,255,085 (Spence) describes a method to adjust the tone reproduction curve of a press or output printer. U.S. Pat. No. 5,255,085 creates a target from the press or desired output proof, benchmarks the characteristics of the proofing device, and discloses a method to generate a lookup table to adjust the dot-gain of the original file to achieve the aim on the proofing device. U.S. Pat. No. 5,293,539 (Spence) adds adaptive process values to interpolate between measured Benchmark and Aim data sets to calibrate the dot-gain tone-scale curve at other screen rulings, screen angles, and dot shapes. Utilizing these techniques to modify the dot-gain curves and hence the tone-scale curves of the proofing device increases the chances for error. The input file and its subsequent components must be available for both RIPs. The same versions of each file and components must be specified. The same fonts must be available for both RIPs. The correct dot-gain curve must be specified at both RIPs. The chances for error to occur increase with each ripping operation, especially when the RIPs are located at separate sites.
Ripping the file twice is also time consuming. Each RIP operation must read the input files, decide where each of the components is to be placed in the output print, convert continuous tone images using the correct dot-gain curve into high resolution halftones, render text and linework, and output a high resolution bitmap which represents the composite image. This is repeated for each color in the output print.
Once a commercial halftone proofer implements dot-gain by modifying the code values being printed through a curve prior to converting the code values into the halftone bitmap with the raster image processor. The dot-gain is only applied to the continuous tone image data and not the line work or text. The dot-gain may be adjusted for each of the primary colors cyan, magenta, yellow, and black. A dot-gain curve may also be specified for spot colors orange, green, red, blue, white, and metallic. A dot-gain curve may also be specified for a
            Equation        ⁢                  ⁢    1    ⁢                  ⁢          Murray        ⁢          -        ⁢          Daives        ⁢                  ⁢          Dot        ⁢                  ⁢          Area        ⁢                  ⁢          Calculation            PercentArea    =                            10                      -            Dtint                          -                  10                      -            Dpaper                                                10                      -            Dsolid                          -                  10                      -            Dpaper                              recipe color which is imaged using a single bitmap in combination of two or more standard colors at unique exposure levels. A dot-gain curve may also be specified for each colorant within a recipe color. In this last case more than one bitmap is used, however the halftone dots are at the same screen ruling, screen angle, and phase, such that each halftone dot in each color substantially overlap.
A typical example is a target curve. Such a target might specify that the 50% cyan halftone should print at 67%, the 25% cyan halftone should print at 35%, and the 75% cyan halftone should print at 80%. A benchmark proof is then run and measured. Dot area is calculated based on measured density using the equation defined by Murray-Davies. Equation 1 is the Murray-Davies equation is defined in ANSI/CGATS.4-1993, 1993, p. 7. A dot-gain adjustment curve is then created to add the correct amount to cyan to achieve the target values at the target inputs. For instance in this example we might find that an output value of 35% was achieved at an input level of 30% in the benchmark proof. Therefore 5% dot-gain at the 25% input level is added to achieve the 35% target. At the 50% level we may find we achieved the target level of 67% at an input level of 57% requiring us to add 7% at the 50% input. At the 75% level we may find we achieved the 80% target at the 76% input requiring 1% dot-gain. In actual practice we may measure the dot-gain in 5% or 10% steps with some additional measurements between 0 to 10% and 90 to 100%. A spline curve is usually fit to the resulting dot-gain curve to provide a table in 1% input increments or less. Smoothing is sometimes performed on the input target and benchmark data to further reduce artifacts in the adjustment process.
Perup Oskofot has shown a software program, which operates on high resolution scans from their scanners. The program takes a binary high-resolution scan of a halftone film and de-screens it to a lower resolution continuous tone image. Typically the scan resolution is 2400 dpi. The resulting continuous tone image may be 8 bits per pixel at 300 dpi resolution. A dot-gain curve is then applied to the de-screened image. The adjusted image is then ripped to a bitmap image at 2400 dpi. This software system was disclosed at Drupa 2000, a tradeshow. One problem with this method is that it requires a re-ripping step. To accomplish this requires a RIP. Plus we need to know what the original halftone screen shape, screen ruling, and screen angle were in order to faithfully reproduce it with the re-ripping step. Another problem is that all RIPs are not the same. There are subtle differences between them such as the method that they use to add noise to hide the quantization affects in screening the image. This means that one RIP may not sufficiently reproduce all the screens that the customer might digitize. Another problem with this method is that it is extremely slow. A small 8×10 inch image at 2400 dpi scanned resolution took more than an hour to process a single color plane.
Additionally, some customers have halftone films, which they would like to use in their digital workflow. These customers scan the film at a high resolution, for example 100 pixels/mm, and quantize each pixel to a binary value. Because the dot-gain is built into the film, there is no method other than de-screening the bitmap file, adding dot-gain, and re-ripping the file, to calibrate the output print. If the original film was made using an optical technique then the dot shape, screen ruling, and screen angle may not be an exact match to a digital RIP. De-screening and re-screening the high resolution scan may not faithfully reproduce the original screens.
Denber et. al. disclose a method of shifting and adding a bitmap image with itself to thin the image displayed in U.S. Pat. No. 5,250,934. Denber discloses a method of setting a bit to an intermediate level if it is diagonally between two active bits using shifting, logical and, and a logical or operation.
U.S. Pat. No. 5,483,351 (Mailloux et al.) discloses using a 4×4 input to a lookup table to determine how to operate on the central 2×2 pixels to implement half bit or full bit dilation and erosion in U.S. Pat. No. 5,483,351. U.S. Pat. No. 5,483,351 has the advantage of knowing some of the surrounding pixels in deciding how to dilate or erode the pixels in the center. Eschbach teaches us in U.S. Pat. No. 5,258,854 how to resize bitmap images in small amounts less than one full bit in size.
Loce et al. discloses logically combining two morphological filter pairs and an original image to create an output image in U.S. Pat. No. 5,680,485. The morphological filters described are erosion filters, one of which has less erosion than desired and the other having more erosion than desired. Logically combining combinations of the original image with the two eroded images provides for a method of obtaining an intermediate result.
Eschbach describes a method of resizing an input bitmap in U.S. Pat. No. 5,208,871, which simulates a scan of an output image from an input bitmap such that the scan resolution is different from the input bitmap. Error diffusion is utilized to quantize the output bitmap into the desired output bit resolution. This example uses error diffusion to spread out the error in the quantization of a multilevel pixel into a reduced number of output states.
U.S. Pat. No. 6,115,140 (Bresler et al.) uses a de-screened version of an original image, and dilated and eroded versions of the original image to select a combination of the original, dilated, and eroded images to effect a dot-gain or tone-scale change in an input bitmap image. U.S. Pat. No. 6,115,140, FIG. 5B shows an original halftone image input into block HI along with an eroded version (HE), and two dilated versions (HD1 and HD2). Then a weight based on de-screened versions of the original halftone (CO), the color corrected original (CI), the eroded original (CE), and the two dilated originals (CD1 and CD2) is calculated. The de-screened images are used to select which of the four halftone images, HI, HE, HD1, and HD2, are transferred into H1 and H2. The weighting function is then used to merge bitmap versions of H1 and H2 together into the tone-scaled output bitmap (HO). How to de-screen is not disclosed, nor exactly how to calculate which bit of H1 and H2 is used to drive the output bit HO. The need to use error diffusion to distribute the error in selecting between H1 and H2 is not mentioned.
In U.S. Pat. No. 6,115,140 dilation is described as growing a single pixel completely around the halftone feature. A second dilation grows two pixels completely around the halftone feature. Similarly erosion subtracts a single pixel completely around the halftone feature.
None of the Bresler et al. references teach how to perform de-screening. Roetling performs de-screening by comparing the number of white and dark pixels within a specified area in U.S. Pat. No. 4,630,125. U.S. Pat. No. 4,630,125 also states that “A partial solution known in the art is to spatially filter the halftone image with a low pass filter.” U.S. Pat. No. 4,630,125 teaches that the spatial filter method is not exact as it tends to blur the original image.
U.S. Pat. No. 5,767,887 issued to Warner et al. discloses using a Raster Image Processor with two lookup tables for dot-gain. One lookup table is recommended for creating a proof. A second lookup table or dot-gain is recommended for making a plate. The image is processed two times through the raster image processor. Warner disclosed imaging the proof and the plate on the same machine, with the same raster image processor. This is not always possible if the proof and the plate are needed in different locations.
Furusawa et al. in U.S. Pat. No. 5,721,625 discloses using a digital filter to filter an input continuous tone image and use the filtered output to select from multiple dot generators or raster image processors. Furusawa selects a dot created using a traditional amplitude modulated screen for areas of the print that contain low frequency information. Furusawa selects a dot created using frequency modulated screens for areas of the print that contain high frequency information. The frequency content of the image is output from the digital filter.